Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    instances of the end-of-text / end-of-turn special token (conversation or sequence boundary markers).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
     todos
    -0.07
     reverse
    -0.07
    -election
    -0.07
     Stein
    -0.07
    杨
    -0.07
    _used
    -0.07
    emed
    -0.07
    ি
    -0.07
     psi
    -0.07
    鲁
    -0.07
    POSITIVE LOGITS
    (Border
    0.07
    รถ
    0.06
     CentOS
    0.06
     trot
    0.06
    BundleOrNil
    0.05
     […
    0.05
    .squareup
    0.05
     fChain
    0.05
    _percent
    0.05
    (Parcel
    0.05
    Act Density 0.044%

    No Known Activations